AI News

Follow the latest AI and automation news

22 Jun 20267 Muharram 1448 AH

Nvidia says its AI data center design runs hotter to use a lot less water

Nvidia has introduced a new liquid-cooled data center design that significantly reduces energy and water consumption. The Rubin generation reference design claims to eliminate massive amounts of power and nearly all water usage, yet it does not address all concerns, including construction impacts and power generation needs. This innovation could reshape future data center construction, but companies must consider the costs associated with implementing it compared to traditional air-cooled systems.

Agents

22 Jun 20267 Muharram 1448 AH

The AI world is getting ‘loopy’

AI agent technology has taken a significant step forward by enabling a group of agents to operate continuously in the background. These agents can perform tasks simultaneously, enhancing operational efficiency. This advancement represents a crucial development in AI, allowing agents to collaborate without the need for constant human intervention. Such autonomous operation enables faster data processing and decision-making. These innovations could lead to substantial improvements in business execution, saving time and resources, and may open new horizons in fields like robotics and automation.

Governance

22 Jun 20267 Muharram 1448 AH

Meta Exposed Data Internally From Its Controversial Employee-Tracking Program

Employees have expressed concerns regarding an initiative to collect their keystroke data for training AI models. This initiative raises questions about privacy and the use of such data. As reliance on AI increases, addressing these concerns is essential to maintain trust between employees and management. The initiative involves collecting precise data on typing patterns, which could enhance AI model performance. However, using this data without clear consent may result in privacy violations. Companies must establish clear policies to protect employee data and promote transparency in how this information is utilized.

💡

AI chipmaker Groq confirms $650M raise, re-staffs after Nvidia’s $20B not-acqui-hire deal

What impact do you think Groq's new investments will have on the AI market?

Business

Micron Technology Invests in Anthropic and Secures Strategic Supply Agreement

22 Jun 20267 Muharram 1448 AH

Business

22 Jun 20267 Muharram 1448 AH

Micron Technology Invests in Anthropic and Secures Strategic Supply Agreement

Micron Technology signed an agreement with Anthropic for memory and storage supply, alongside a strategic investment in its latest funding round. Companies are racing to secure essential components for increasingly costly data centers, driven by rising demand for high-bandwidth memory. Tom Brown, CEO of Anthropic, emphasized that memory and storage are crucial for the efficiency of AI model training. Anthropic has also secured deals with companies like Corwave, Broadcom, and SpaceX to enhance its computational capabilities.

Models

Google makes Interactions API the default interface for Gemini models and agents

22 Jun 20267 Muharram 1448 AH

Models

22 Jun 20267 Muharram 1448 AH

Google makes Interactions API the default interface for Gemini models and agents

Google DeepMind has announced that the Interactions API is now the default interface for its Gemini models. This new interface replaces the old generateContent API and utilizes a simplified schema with typed steps instead of role-based structures. Future agent features will only be delivered through this API, indicating a significant shift in how models will interact with users.

Governance

Three things to watch amid Anthropic’s latest feud with the government

22 Jun 20267 Muharram 1448 AH

Governance

22 Jun 20267 Muharram 1448 AH

Three things to watch amid Anthropic’s latest feud with the government

In April, Anthropic announced its AI model Mythos, deemed a cybersecurity threat. Following this, it released a modified version called Fable, but the US government imposed export controls due to safety concerns. This government intervention raises questions about reliance on American AI companies. Amazon CEO Andy Jassy flagged Fable as dangerous, prompting swift government action. This has sparked reactions in Europe, with politicians urging the development of local AI models. The situation reflects a shift in how European nations view American AI technologies.

Agents

Amazon is testing Alexa+ in India with Hindi support

22 Jun 20267 Muharram 1448 AH

Agents

22 Jun 20267 Muharram 1448 AH

Amazon is testing Alexa+ in India with Hindi support

Amazon is set to expand its conversational AI assistant Alexa+ into India, inviting users to test a Hindi-language version. This initiative aims to strengthen Amazon's presence in the rapidly growing AI market in India, catering to Hindi-speaking users. The new version of Alexa+ is designed to meet the specific needs of this demographic. The expansion of Alexa+ in India presents a significant opportunity for Amazon to enhance user engagement with its technologies, potentially leading to increased sales and growth in this emerging market.

Business

Anthropic and Micron want to co-design AI memory architecture

22 Jun 20267 Muharram 1448 AH

Business

22 Jun 20267 Muharram 1448 AH

Anthropic and Micron want to co-design AI memory architecture

Micron has announced its investment in Anthropic's Series H funding round, securing a multi-year deal to supply memory for Claude's infrastructure. Tom Brown, co-founder of Anthropic, emphasizes that memory is critical for training and running Claude, highlighting the importance of this collaboration in enhancing AI capabilities. This move comes as Micron's stock has surged over tenfold in the past year, raising questions about the impact of such circular deals on the market.

Governance

OpenAI Launches Full-Scale Effort to Patch Open-Source Bugs as It Takes on Anthropic’s Mythos

22 Jun 20267 Muharram 1448 AH

Governance

22 Jun 20267 Muharram 1448 AH

OpenAI Launches Full-Scale Effort to Patch Open-Source Bugs as It Takes on Anthropic’s Mythos

OpenAI has unveiled the enhanced version of GPT-5.5-Cyber, aimed at bolstering the cybersecurity capabilities of AI models. This initiative comes amid growing concerns over security vulnerabilities in open-source software. Additionally, the company launched the "Patch the Planet" initiative, focusing on fixing bugs in open-source software. Through this initiative, OpenAI aims to tackle the security challenges faced by existing models, as vulnerabilities pose an increasing threat in the AI landscape.